The Double Enclosure of Co- Produced Knowledge: Moving Towards Open-Access Scholarly Infrastructure and then All Good?

Full text

Big Data / Smart Data
Chapter 1

2 Frontiers in African Digital Research
University of Bayreuth African Studies Online (10)

The Double Enclosure of Co-
Produced Knowledge
Moving Towards Open-Access Scholarly
Infrastructure and then All Good?
Stefan Ouma, University of Bayreuth
1
1 Introduction
This paper engages with various types of knowledge infrastructures that have proliferated in the
world of academic research and publishing over the past two decades, a process that has been
thoroughly reshaped by the power of digitization. One way to go about the research and
publication process is to conceive of it as a largely technical affair. Such a stance has been
convincingly refuted by numerous works since at least the 1970s, which have exposed the social
and political situatedness of research, writing and publishing. Not only the personal is political,
but research is, too! These critical insights can be further extended to incorporate the very
infrastructures of publishing, which, today are largely digital infrastructures. Often, these are part
of the ever-growing ‘ecosystems’ of commercial publishing empires. Imperial analogies seem
justified given the sheer size of large publishers and their value extraction strategies, which are
fundamentally built on capitalizing on the benefits from free or low-paid labour (authors,
reviewers, editors, technical support staff, often working from Global South countries) and from
largely publicly funded-research (Monbiot 2018).
1
I thank Prof. Tanu Biswas (University of Stavanger) for earlier comments on this paper, as well as one anonymous
reviewer.

4 Frontiers in African Digital Research
University of Bayreuth African Studies Online (10)
This contribution proceeds as follows. The first section (3) problematizes a purely technical and
assumingly apolitical take on the production and ‘dissemination’ of academic knowledge. It shows
how the sociological concept of embeddedness can be utilized to engage with the multiple
relationalities underpinning the ‘traditional’ production of research knowledge, which is already
characterized by several moments of scholarly enclosure. In the next section (2), I show how since
the 1990s, scholarly enclosures are increasingly being followed by a second round of enclosures
through the corporately controlled process of ‘knowledge dissemination’. This process has
become governed largely by a few commercial publishers, which have also acquired controlling
stakes in other nodes of the ‘knowledge value chain’, including its very infrastructure. I then (3)
differentiate such enclosed knowledge infrastructures from open-access (OA) scholarly
infrastructures, which are lately being promoted as an alternative to the corporate publishing
model and as a more general way of how research data should be stored, made accessible, used
and re-used. Despite the promises read into such infrastructures, there is also a range of risks
attached to their uncritical promotion. I shall outline these risks by drawing on perspectives from
critical data studies. I close (4) by finding that a research agenda poised to reconfigure ‘African
Studies’, to which this intervention seeks to contribute in the spirit of the Cluster of Excellence
“Africa Multiple” at the University of Bayreuth
2
, must also reconfigure the way that contemporary
knowledge infrastructures work, as well as what is being fed into them.
2 From disembedded to embedded knowledge production
Even though scores of papers and books have made the point that knowledge production is
socially and politically situated (for a classic, see Haraway 1988), such views are still finding a
hard time with many social and natural scientists who still think of research as an “an innocent
pursuit of knowledge” (Ndlovu-Gatsheni 2017). Self-assuring statements on being objective,
neutral, apolitical or deploying a merely positive approach to the world are indicative of this
stance (Friedman 1953) and resurface from time to time (for a recent example, see Basedau
2020). Yet they are not as apolitical as they appear at first sight, because they are usually
mobilized to deem those who are committed to a version of theory that challenges the status quo
more radically as being unscientific, normative, or ideological. For these scholars, it is also quite
easy to conceive of the knowledge production and dissemination process in purely technical terms
(see Figure 1).
Those who claim that all research is socially embedded, including this author, posit that it is
shaped by structural features, i.e. a historically grown coloniality of being, knowledge and power
(Ndlovu-Gatsheni 2020; Kellecioglu 2020; Smith 2012) that shapes how scholarly knowledge is
produced and ‘distributed’. This can also be conceived as a specific kind of structural
embeddedness, a term originally coined by economic sociologist Mark Granovetter (1985) and
more recently reworked by (Hess 2004). This structural embeddedness also shines through in the
relational embeddedness of knowledge production, which signifies the social networks into which
knowledge production is embedded, variously comprising of funders, research participants,
donor- and recipient-researchers, research assistants, and publishers/editors of books and
journals.
2
See https://www.africamultiple.uni-bayreuth.de/pool/dokumente/Africa-Multiple_Overview-1.pdf (7 September
2021).

The Double Enclosure of Co-Produced Knowledge 5
University of Bayreuth African Studies Online (10)
Contrary to what the term network suggests, the social structures of knowledge production in
question are not heterarchical (Deane and Stevano 2016; Marchais et al. 2020). Among many
others, Judith Buhenwa Nshobole (2020) emphasizes the often-asymmetrical relationship
between either senior scholars (who often serve as principal investigators and money holders)
and junior scholars (PhD students and postdocs) in the North and South alike and between
Northern scholars and their Southern counterparts. In other words, knowledge co-production
3
is
co-structured by the intersectional parameters of gender/sex, race, seniority, institutional
affiliation, language, and material endowment, all of which have a history and geography to them.
Much of the unevenness in the global publishing landscape is a product of the interplay between
the specific structural and relational embeddedness of knowledge production (see for example
Addis and Villa 2003 for economics; Bański and Ferenc 2013; Schurr et al. 2020; Jazeel 2016 for
3
I am speaking of co-production here because the knowledge produced by academics could not have been produced
without the help of others. As I have argued elsewhere, “[k]nowledge production must always be conceived of as
situated – as being dependent on the position of the researcher in local/translocal, materially entangled, and power‐
laden social relations. What is often staged as an act of autonomous knowledge production by realists is de facto an
intersectional, often asymmetrically structured co-generation of knowledge” (Ouma 2015: 84). Research participants,
field assistants, co-researchers, and prior accumulated knowledge and experiences made by the researcher are firmly
part of that co-production (Schwartz-Shea and Yanow 2012: 80).
Figure 1: Figure 1: The academic knowledge production process and the traditional role of
publishers. Source: Redrawn from Posada and Chen 2018.

6 Frontiers in African Digital Research
University of Bayreuth African Studies Online (10)
geography; Fox et al. 2016 for ecology and evolution). Lastly, territorial embeddedness shapes the
production and dissemination of knowledge, and adds a further piece to the unevenness.
Researchers need to stick to the laws of the country they are working in while carrying out
research. Certain type of data may or may not be in the public domain, depending on the
regulations of the country (property data would be one prominent example). Furthermore, much
of the dissemination of knowledge via journals and books is governed under global copyright
arrangements that favour corporate publishing houses and can quickly cause trouble for those
who have moved against the prevailing intellectual property regime (for a recent example, see
Banka 2020).
Once we have adopted an embeddedness view of the knowledge production and dissemination
process, several issues come to the fore. First, this process happens in a field of forces, whereby
issues of power, context, subjectivity/positionality, language, and practice are important entry
points for a debate on how research works across all disciplines. Figure 2 untangles these entry
points even though in practice, they cannot be separated (see Figure 2, e.g., all knowledge is
contextual, produced via practices, shaped by power relations. language and positionality). On the
‘input side’, what knowledge can be produced and how it is being produced is further shaped by
the policies and politics of funding. For an illustrative controversy on the latter issue, see the push
of the later Trump administration to re-organize the funding going to Geography as a diverse
discipline that covers both natural and social sciences into the funding docket “Human-
Environment and Geographical Sciences Program (HEGS; previously known as Geography and
Spatial Sciences) (Hamilton 2020). This move was deliberately designed to marginalize
qualitative constructivist, political economy or otherwise critical geographical work. While neo-
Positivism may be quickly blamed as the force at work here, it is clear that this was part of a larger
attack on critical voices by the Trump administration, which also included attacks on otherwise
largely neo-Positivist fields such as climate change science.
On the ‘output side’, the knowledge produced is packaged into publishable formats, most
important of which are still journal articles and books (however, datasets and codes have become
new important outputs). These are usually only accessible at a cost and are often hidden in the
abysses of the publishing industry. This contrasts sharply with blog posts, open-eds or other forms
of public engagement, which are meant to bring scholarly work in dialogue with society.
Second, both the production of small and big data, can be placed within this field of forces. While
traditionally, humanities and social science research has been deemed to produce ‘small data’,
recent technological advances across fields relevant for the humanities, social sciences and
natural sciences increasingly produce, or at least make use, of ‘big data’. The latter is often very
large, covers entire populations, has a tight resolution and strong indexicality, offers strong
insights into relationality (e.g., social network data), has a fast velocity, covers a wide variety of
variables, and offers a high flexibility and scalability (Kitchin and Lauriault 2015). Big data
includes data that researchers use, but was produced by others, such as the data that can be
extracted from satellite images, or social network data (e.g., Tweets) or certain forms of
government data. Big data can also become big when traditional data sets, including images or
field notes, are digitized and become part of a larger platform that allows for new ways of
researching, and makes the content open “to the application of big data analytics such as data

The Double Enclosure of Co-Produced Knowledge 7
University of Bayreuth African Studies Online (10)
mining, pattern recognition, data visualization, statistics and modelling” (Kitchin and Lauriault
2019: 86).
Figure 2: An embedded model of scholarly knowledge production and dissemination.
Source: Author’s design, with inspiration from Zienkowski 2017.
‘Big data’ contrasts sharply with traditional ‘small data’, which, besides some “exceptions, such as
population censuses or meteorological data collection”, were “typically characterized by limited
volume, sampled collection, small geographic extent, and narrow variety and framing” (Kitchin
and Lauriault 2019: 84).
Third, and central to this contribution, the interplay of forces in the field of knowledge production
as we know it, leads to several forms of scholarly enframing:
 a theoretical enframing, whereby scholars develop a frame that allows them
to organize the knowledge co-produced in a certain way while excluding other
theoretical framings, including those that might be offered by the research
participants themselves.

8 Frontiers in African Digital Research
University of Bayreuth African Studies Online (10)
 a detachment of the knowledge produced from its larger context, which
transforms something that was co-produced through interpersonal social
relations into impersonal ‘data’;
 an exclusion of all those players, who may have played a crucial role during
the research process, including other scholars, survey/interview
respondents, research assistants and enumerators.
While we could be satisfied with describing these processes as acts of framing – “the tracing of a
boundary between relationships and events which are internalized and included in a decision or,
by contrast, externalized and excluded from it” (Callon 1998: 15), they can be likewise described
as scholarly enclosures. The enclosed product of knowledge work can be easily commodified
through market transactions precisely because of previous acts of enframing. Scholarly
enclosures here refers to “efforts to secure and reinforce specific scholastic territories” (Biltoft
2019: 232). While deeply enmeshed with the logic of ‘modern’ knowledge production, such
enclosures have further accelerated over the past 20 years or so due to the operational logic of
the neoliberal academy (‘publish or perish’; ‘the rank and yank academy’, Berg 2012):
Academics these days are mainly trained to write for each other and not the general
reader. It wasn’t always like this but over the years, there has been a kind of scholarly
enclosure, especially in the West […]. A kind of scholarly enclosure has advanced as
academics are encouraged to address whatever conversation seems to be in vogue in
a particular moment, and this is often the one that others can’t understand, and all of
this becomes further validated through the inwardly looking practices we perpetuate
of recognition, citation and promotion. (Shah 2019)
What becomes clear here is that in relation to scholarly work, enclosure can be conceived as a
product of specific knowledge practices as well as of commodification processes. Both are
intertwined. While most researchers cannot directly transform their research outputs into ready
cash (though some do, e.g. via patents or lucrative book contracts), scholarly enclosures still help
them advancing their careers, while those who have crucially contributed to the underlying
research are lucky if they are mentioned in the acknowledgements section of a paper, book or
dataset. Any project poised to reconfigure African studies, or any other field of social and natural
science inquiry, must firmly reflect on this first act of enclosure and overcome it in favour of a
model of “socially just publishing” (Batterbury 2017). This call is pertinent because what is
problematized here has been the classic mode of how research has operated over centuries, which
was further entrenched by the publish-or-perish ethos of the neoliberal academy:
Decolonising methodology must begin with unmasking the modern world system and
the global order as the broader context from which re-search and methodology are
cascading and are influenced. It also means acknowledging and recognising its
dirtiness. Our present crisis is that we continued to use re-search methods that are
not fundamentally different from before. (Ndlovu-Gatsheni 2017)

The Double Enclosure of Co-Produced Knowledge 9
University of Bayreuth African Studies Online (10)
3 The corporate enclosure of research knowledge
Scholarly enclosures are a long-standing product of the academic knowledge co-production
process, which too often masquerades as ‘innocent research’. However, the emphasis on historical
continuity can easily make us forget the novel elements that co-structure the process of academic
knowledge production and circulation. After this first, ‘traditional’ round of academic enclosure,
a more recent and more direct form sets in
4
: the publication of academic work via the oligopolistic
structures of the publishing world. Today a few “megapublishers” (Peekhaus 2012: 577) – a fairly
recent development (Berg 2012; Buranyi 2017; Larivière et al. 2015), exert control over several
nodes in the ‘knowledge value chain’ rather than merely having stakes in the dissemination of
journal articles and books. In other words, rephrasing Brian Larkin, one of the most prominent
critical thinkers on infrastructures, these publishers increasingly control the architecture for the
circulation of scholarly knowledge, literally providing the infrastructure of the publishing world
of the early 21st century, generating “the ambient environment of everyday [academic] life”
(Larkin 2013: 328) – including for the making of academic careers (e.g., via metrics such as the
Impact Factor).
Figure 3 highlights how the publication landscape has become increasingly concentrated since the
mid-1990s, resulting in a shift from society- or university-owned and small-publishing houses (all
deemed “small publishers” in figure 3) to “big publishers”. This shift has been most striking for
the natural and medical sciences, which promise larger turnover volumes due to their sheer size.
The result of this secular development has been that by 2015, the five large publishers (RELX
Group including Elsevier, Wiley-Blackwell, Springer, Taylor & Francis) published around 70% of
social science articles and 20% in the humanities (Larivière et al. 2015). Their oligopolistic market
power has manifested itself “in skyrocketing journal subscription costs and a drastic lockdown of
content through strict application of copyright and licensing restrictions” (Peekhaus 2012: 577).
Berg (2012) highlights the consequences of the pay- and copyright wall for both relatively well-
endowed institutions in the Global North and their less endowed counterparts in the Global South:
This profit-taking has significant impacts on the relatively well-to-do academic
libraries of major research universities in the global North, many of which have had
to reduce their serials subscriptions in order to meet their budgets […]. At the same
time, this profit-taking also operates to prevent scientific knowledge circulation,
especially in the Global South […]. (Berg 2012: 261)
4
As Batterbury (2017: 175) argues, for “some, it is perfectly acceptable to cede author copyright to companies that
prepare and sell them, thereby losing ownership and management of that intellectual property.” For him and others,
including this author, research funded with public money should be a public good (Monbiot 2018). I would extend that
for both social justice and ecological reasons, generous intellectual property protections must also be dismantled in
favour of true OA regimes of knowledge (Fuchs and Sandoval 2013; Raworth 2017).

10 Frontiers in African Digital Research
University of Bayreuth African Studies Online (10)
Figure 3: Number of journals changing from small to big publishers, and big to small
publishers per year of change in the Natural and Medical Sciences and Social Sciences &
Humanities. Source: Redrawn from Larivière et al. 2015.
5
Oligopolistic competition in publishing has resulted in staggering operating profits and profit
margins for the mega-publishers. Larivière et al. (2015) document this for the case of the
publishing giant Elsevier (see Figure 4), but older data on other large publishers suggests similar
levels of profitability (Fuchs and Sandoval 2013).
5
The category “from small to big publisher” includes journals that moved between small publishers (i.e. any publisher
but the five) and big publishers (i.e. those top five that are presented in Figure 3). Personal information provided by
Vincent Larivière, 09 July2021.

The Double Enclosure of Co-Produced Knowledge 11
University of Bayreuth African Studies Online (10)
Figure 4: Operating profits (million USD) and profit margin of Reed-Elsevier as a whole (A)
and of its Scientific, Technical & Medical division (B), 1991–2013. Source: Redrawn from
Larivière et al. 2015.
As indicated above, publishers also increasingly try to acquire control over the very infrastructure
of academic knowledge production, circulation and evaluation, which, again, can be well
documented for the case of Elsevier (see Figure 5). Thus, they increasingly pose as digital
landlords (Sadowski 2020) or ‘knowledge rentiers’, who generate rents by owning the very
infrastructure that allows researchers to collaborate, offer their scholarly ‘crops’, or lets users
purchase them and evaluate their quality (Posada and Chen 2018). Rentiers, here, are defined in
line with Sadowski (2020: 565) drawing on Felli, “by their ‘ownership of the access to a condition
of production’ (Felli 2014: 269) and their ability to derive income (rent) from access to assets.”
Because in general, access to the sources of rent is either limited by nature (e.g. in the case of land)
and/or by state regulation (e.g. in the case of intellectual property), those who have property
rights over the sources are usually able to generate at least a part of their income from non-
productive activities. Income without corresponding investment often leads to above-market rate
of returns. Since publishers do usually process the ‘inputs’ (= research) paid for by the state
(Monbiot 2018), and can rely on the largely free labour of editors, editorial board members,
authors and reviewers, they are able to extract significant amounts of surplus value (Berg 2012).
Historically, infrastructure has played a central role in extending imperial value relations (Cowen
2018), and the quest of mega-publishers to accumulate ever more infrastructural power, must be

12 Frontiers in African Digital Research
University of Bayreuth African Studies Online (10)
read in this light. Infrastructure is things, but “also the relation between things” (Larkin 2013:
329).
Figure 5: The publishing ecosystem of Elsevier. Source: Redrawn from Posada and Chen
(2018).
4 The Promises and Perils of OA Data Infrastructures
As Batterbury (2017: 175) argues, for “some, it is perfectly acceptable to cede author copyright to
companies that prepare and sell them, thereby losing ownership and management of that
intellectual property.” For him and others, including this author, research funded with public
money should be a public good. It should be added that both social justice and ecological reasons,
generous intellectual property protections must also be dismantled in favour of true OA regimes
of knowledge (Fuchs and Sandoval 2013; Raworth 2017).
As part of an “academic spring”, governments, research councils, librarians, university leaders,
academics and journalists have come out to criticize the corporate publishing model and called
for its overhaul. Corporate publishers have reacted by increasingly offering OA models. These
were able, often supported by governments, to push for a ‘gold OA model’ (Fuchs and Sandoval
2013) which requires authors to pay still rather high article processing charges (APC), while
making content available to readers immediately and at no cost. As Vega et al. (2021) show, this
model is not the dominant one in the global publishing landscape, even if publishers are quick to

The Double Enclosure of Co-Produced Knowledge 13
University of Bayreuth African Studies Online (10)
invoke this impression (Fuchs and Sandoval 2013)
6
. Paper repositories (so-called ‘green OA’)
hosted by universities or other ways to make research findings accessible (e.g., via blogs or
platforms such as THE CONVERSATION) are supplementing the paper-focused OA drive. Ironically,
a prominent definition of OA as “digital, online, free of charge, and free of most copyright and
licensing restrictions” (Suber 2013, in Kitchin et al. 2015: 665), seeking to demolish both “price
barriers (subscriptions, licensing fees, pay-per-view fees) and permission barriers (most
copyright and licensing restrictions)” (ibid.), obscures the significant differences between for-
profit and not-for-profit OA models (which are non-corporate) (Fuchs and Sandoval 2013).
There are a variety of OA models
7
, including assumingly credible publishers such as MDPI that
have become more predatory in their business practices (Crosetto 2021) or OA models that
neither charge authors, nor readers, nor institutions and have maintained a high level of academic
integrity at the same time (e.g., the Journal of Political Ecology). Many, however, see the future of
producing and disseminating research in OA digital data repositories (Kitchen et al. 2015), a.k.a
“Open Scholarly Infrastructures” (Bilder et al. 2015). There are significant differences in how OA
data repositories can be organized and funded, but as a rule of thumb these can be largely
considered as OA infrastructures meant to reshape the way knowledge is produced, used and
reused, with proponents making both scientific and financial arguments for why such
infrastructures should be the future of research (Kitchin et al. 2015; see Table 1).
6
In Vega et al.’s data base, 72% percent of all journals listed do not attract APC.
7
For current trends data on access journals, see Vega et al. (2020) and European Commission (2021a).

14 Frontiers in African Digital Research
University of Bayreuth African Studies Online (10)
Table 1: Benefits of data repositories/infrastructures, source: Kitchin et al. 2015: 666.
Direct benefits
Indirect benefits (costs avoided)
New research opportunities
Scholarly communication/access to data
Re-purposing and re-use of data
Increasing research productivity
Stimulating new networks/collaborations
Data available for teaching and student projects
Knowledge transfer to industry
Improves skills base
Increasing productivity/economic growth
Verification of research/research integrity
Fulfilling mandate(s)
No re-creation/duplication of data
No loss of future research opportunities
Lower future preservation costs
Re-purposing data for new audiences
Re-purposing methodologies
Use by new audiences
Protecting return on earlier investment
Tools and standards have potential to increase
data quality
Reduces ad hoc queries concerning data
Short-term benefits
Long-term benefits
Value to current researcher and students
No data lost from researcher turnover
Widens access where costs are prohibitive for
researchers/institutions
Short-term re-use of well-curated data
Secure storage for data-intensive research
Availability of data underpinning publications
Secures value to future researchers and students
Adds value over time as collection grows and
develops critical mass
Increases speed of research and time to realise
impacts
Stimulates new research questions, especially
relating to linked and derived data
Private benefits
Public benefits
Benefits to sponsors/funders of research/archive
Benefits to researchers and institutions
Fulfils grant obligations
Increased visibility/citations
Commercialising research
Input for future research
Motivating new research
Catalysing new companies and high-skills
employment
Transparency in research funding
Over the past years, a number of these infrastructures have been launched, benefiting from “new
archiving technologies, a massive expansion in digital data storage, data standards, and open
science and open data movements” (Kitchin and Lauriault 2019: 86). Usually, these
infrastructures are operated at national or regional scales. For instance, the European Union
published its first position papers on open science and data in 2012 (Kitchin et al. 2015). Since
then, it has developed an Open Science Policy, centring around eight ambitions. Among others,
this includes calls for Open Data, setting up the European Open Science Cloud (EOSC), developing
a new generation of metrics to measure impact, and strengthening of Citizens Science (European
Commission 2021b).
8
However, it is more appropriate to call such infrastructures ‘living digital
archives’ in order to underline that these not only combine heterogeneous, searchable data
8
I cannot comment in detail on this initiative from the perspective of ‘socially just publishing’. Suffice it to say, the
European Commission has largely failed to offer a more radical and reflexive imagination on how the future of
knowledge production and circulation could look like due ongoing corporate influence (Elsevier has been a major
subcontractor of the project).

The Double Enclosure of Co-Produced Knowledge 15
University of Bayreuth African Studies Online (10)
related to various dimensions of the knowledge co-production process, but also that they need
initial and ongoing social and material investments, including curation:
Archives are formal collections of data that are actively structured, curated and
documented, are accompanied by appropriate metadata, and where preservation,
access and discoverability are integrated into technological systems and institutions
designed to last the test of time (Lauriault et al. 2013). Archives explicitly seek to be
long term endeavours, preserving the full record set—data, metadata and associated
documentation—for future reuse. (Kitchin and Lauriault 2015)
For sure, big data and data infrastructures in the form of repositories or even archives offer
unseen and welcome possibilities, especially if curated as ‘open access’ This notwithstanding,
setting up such archives is fraught with all sorts of financial, managerial and technical challenges
(for a good overview, see Kitchin et al. 2015). Another way to put it is that there are challenges
related to “running the infrastructure (governance), funding it (sustainability), […] preserving
community ownership of it (insurance)”, and implementing it (Bilder et al. 2015).
Moreover, there is the risk of eclipsing the knowledge politics underpinning the call for digital
data repositories. One of those risks is to subject small data to a big data epistemology, as for
instance the European Commission does, when it argues that open data facilitates the replication
of research, which is usually not a quality criteria for most qualitative research. Such a stance
clearly risks that, “knowledges that are not so easily encapsulated within big data frameworks
might become devalued” (Graham and Shelton 2013: 257). Other associated risks include:
 the concentration of power in the hands of new gate keepers (‘data scientists’)
(Donati and Woolston 2017).
 decontextualization, that is when one piece of data becomes just one element
in a giant data heap.
 that sound and engaged secondary research gives way to quick data mining.
 that people assume that large (quantitative) data is the “end of theory”
(Anderson 2008).
This is why data infrastructures, as aggregators of data through which this data becomes big (of
course, it can also further aggregate data that already was big), must be subjected to the same
critical gaze as ‘born-big data’ itself. There are promises and perils of both big data and big data
infrastructures (Graham and Shelton 2013), regardless of whether they are OA or not. For critical
data scholars such as Dalton and Thatcher (2014: n.p.), among others this includes accepting that
big is not everything, that big data is never raw or technologically innocent, as well as the need to
reflect on the crucial questions such as “who controls ‘big data,’ its production and its analysis?
What motives and imperatives drive their work?” and who the subjects of ‘big data’ are and what
knowledges are they producing?
On top, even if one has solved the technical, managerial and financial challenges, and successfully
tackled the epistemological risks just outlined, it does not mean that an archive magically emerges.
As Stuart Hall noted in his classic definition of the knowledge archive,

16 Frontiers in African Digital Research
University of Bayreuth African Studies Online (10)
Constituting an archive represents a significant moment, on which we need to reflect
with care. It occurs at that moment when a relatively random collection of works,
whose movement appears simply to be propelled from one creative production to the
next, is at the point of becoming something more ordered and considered: an object
of reflection and debate. (Hall 2001: 89)
9
While one can barely expect Hall’s idea of the archive to materialize in national-scale initiatives,
this is different for more institution-based archives, where it is possible to craft the intellectual
philosophy of an archive in Hall’s sense into its technicalities - if only those involved are willing
do so. Kitchin and Lauriault (2015) are right when they argue that realizing the idea of
comprehensive digital data repositories would need a cultural change among researchers, but so
would overcoming of the classic scholarly enclosures highlighted above. A common, more radical
philosophy underpinning a given digital archive should also further be extended to an
engagement with the “deep coloniality” (Kellecioglu 2020) that shapes why and how research is
conducted in the very first place, and document the steps that were taken to produce and circulate
knowledge in a substantially different way. Suffice it to say, one can have quite a radical take on
the archive, without undoing the “dirty history” of research (Ndlovu-Gatsheni 2017; Smith 2012).
In fact, this risk should be of significant concern when revisiting the history of ‘techno-
solutionism’. The language of a technological fix used to change undesirable states of the world
has lent itself too often too well to only a superficial critique of what is to be changed. One also
should be wary of the fact that even projects that have aimed at explicitly undoing research and
its dirty history have struggled to do so in practice (Marchais et al. 2020).
Another risk with the promotion of digital data infrastructures is that nothing speaks against
corporate publishers further appropriating the idea (Bilder et al. 2015), especially when
considering the imperial qualities of ‘infrastructure’ (Cowen 2018). In fact, they have already done
so, as I outlined above, and as platforms such as Sage Data Planet, the Sage Methods Platform or
ResearchGate and Academia.edu show. Years of infrastructural thinking on the side of major
publishers, reinforced by the platform gaze of the tech economy, make it relatively easy to turn
accumulated published knowledge and associated services into a platform product. The fact that
large publishers managed to appropriate OA via APC schemes, largely with the consent approval
of academics and states, serves as a warning here.
Reconfiguring African Studies requires a proper engagement with the politics and political
economy of a shifting publishing landscape. The framework text that outlines the research
programme of the Cluster of Excellence “Africa Multiple” hints at this, and the plans to create a
digital research environment that makes the research produced more accessible can be read as a
response to the problem of double enclosures in the world of digitally mediated corporate
publishing.
10
This contribution serves a reminder that more work needs to be done in order to
9
This is why the distinction between the “anarchic totality of the repository” and “the curated and ordered logic of the
analogue archive” seems useful. I thank the anonymous reviewer for this point. Of course, the curation of the analogue
archive can also be translated into the digital archive.
10
See https://www.africamultiple.uni-bayreuth.de/en/Digital-Solutions/index.html (7 July2021).

The Double Enclosure of Co-Produced Knowledge 17
University of Bayreuth African Studies Online (10)
come to terms with the question of how knowledge is co-produced, and how that knowledge is
stored and made accessible to the broader public in a socially just manner.
5 References
Addis, Elisabetta, and Paola Villa. 2003. The Editorial Boards of Italian Economics Journals:
Women, Gender, and Social Networking. Feminist Economics 9 (1): 75–91. doi:
10.1080/1354570032000057062.
Anderson, Chris. 2008. The editorial boards of Italian economics journals: Women, gender, and
social networking. Wired Magazine.
https://www.uvm.edu/pdodds/files/papers/others/2008/anderson2008a.pdf (last
accessed: 25 August 2021).
Banka, Richa. 2020. Delhi HC asks Libgen, Sci-Hub to stop uploading articles as they face copyright
infringement charges. Hindustan Times, December 24.
https://www.hindustantimes.com/india-news/delhi-hc-asks-libgen-sci-hub-to-stop-
uploading-articles-as-they-face-copyright-infringement-charges/story-
cRWCB1sGs1yMqR3TCpuvmL.html (last accessed: 25 August 2021).
Bański, Jerzy, and Mariola Ferenc. 2013. “International” or “Anglo-American” journals of
geography? Geoforum 45:285–295. doi: 10.1016/j.geoforum.2012.11.016.Basedau, Matthias.
2020. Rethinking African Studies: Four Challenges and the Case for Comparative African
Studies. Africa Spectrum 55 (2): 194–206. doi: 10.1177/0002039720945328.
Batterbury, Simon. 2017. Socially just publishing: implications for geographers and their journals.
Fennia - International Journal of Geography 195 (2): 175–181. doi: 10.11143/fennia.66910.
Berg, Lawrence D. (2012). Knowledge enclosure, accumulation by dispossession, and the
academic publishing industry. Political Geography 31 (5): 260–262. DOI:
10.1016/j.polgeo.2012.05.011.
Bilder, Geoffrey, Jennifer Lin, and Cameron Neylon. 2015. Principles for Open Scholarly
Infrastructures-v1. https://cameronneylon.net/blog/principles-for-open-scholarly-
infrastructures/ (last accessed: 25 August 2021).
Biltoft, Carolyn N. 2019. Against Scholarly Enclosures: Reconsidering the Art and Economics of
Review. Capitalism: A Journal of History and Economics 1 (1): 231–240. doi:
10.1353/cap.2019.0000.
Buranyi, Stephen. 2017. Is the staggeringly profitable business of scientific publishing bad for
science?, June 27. https://www.theguardian.com/science/2017/jun/27/profitable-business-
scientific-publishing-bad-for-science (last accessed: 25 August 2021).
Callon, Michel. 1998. Introduction: The Embeddedness of Economic Markets in Economics. The
Sociological Review 46 (1_suppl): 1–57. doi: 10.1111/j.1467-954X.1998.tb03468.x.
Cowen, Deborah. 2018. The jurisdiction of infrastructure: Circulation and Canadian settler
colonialism. The Funambulist (17): 14–19.

18 Frontiers in African Digital Research
University of Bayreuth African Studies Online (10)
Crosetto, Paolo. 2021. Is MDPI a predatory publisher?
https://paolocrosetto.wordpress.com/2021/04/12/is-mdpi-a-predatory-publisher/ (last
accessed 25 August 2021).
Dalton, Craig, and Jim Thatcher. 2014. What does a critical data studies like, and why do we care?
Society+Space. https://www.societyandspace.org/articles/what-does-a-critical-data-studies-
look-like-and-why-do-we-care (last accessed 25 August 2021).
Deane, Kevin, and Sara Stevano. 2016. Towards a political economy of the use of research
assistants: reflections from fieldwork in Tanzania and Mozambique: Reflections from
fieldwork in Tanzania and Mozambique. Qualitative Research 16 (2): 213–228. doi:
10.1177/1468794115578776.
Donati, Gaia, and Chris Woolston. 2017. Data domination. Nature (548): 612–614.
European Commission. 2021a. Trends for open access to publications.
https://ec.europa.eu/info/research-and-innovation/strategy/goals-research-and-
innovation-policy/open-science/open-science-monitor/trends-open-access-publications_en
(last accessed 25 August 2021).
European Commission. 2020b. Open Science. https://ec.europa.eu/info/research-and-
innovation/strategy/strategy-2020-2024/our-digital-future/open-science_en (last accessed
25 August 2021).
Felli, Romain. 2014: On Climate Rent. In: Historical Materialism 22 (3-4): 251–280. DOI:
10.1163/1569206X-12341368.
Fox, Charles W., C. Sean Burns, and Jennifer A. Meyer. 2016. Editor and reviewer gender influence
the peer review process but not peer review outcomes at an ecology journal. Functional Ecology
30 (1): 140–153. doi: 10.1111/1365-2435.12529.
Friedman, Milton. 1953. Essays in positive economics. Chicago: Univ. of Chicago Press.
Fuchs, Christian, and Marisol Sandoval. 2013. The Diamond Model of Open Access Publishing: Why
Policy Makers, Scholars, Universities, Libraries, Labour Unions and the Publishing World Need
to Take Non-Commercial, Non-Profit Open Access Serious. tripleC: Communication, Capitalism
& Critique. Open Access Journal for a Global Sustainable Information Society 11 (2): 428–443.
doi: 10.31269/triplec.v11i2.502.
Graham, Mark, and Taylor Shelton. 2013. Geography and the future of big data, big data and the
future of geography. Dialogues in Human Geography 3 (3): 255–261. doi:
10.1177/2043820613513121.
Granovetter, Marc. 1985. Economic action and social structure: The problem of embeddedness.
The American Journal of Sociology 91 (3): 481–510.
Hall, Stuart. 2001. Constituting an archive. Third Text 15 (54): 89–92. doi:
10.1080/09528820108576903.
Hamilton, Aretina R. 2020. The white unseen: On white supremacy and dangerous entanglements
in geography. Dialogues in Human Geography 10 (3): 299–303.

The Double Enclosure of Co-Produced Knowledge 19
University of Bayreuth African Studies Online (10)
Haraway, Donna. 1988. Situated Knowledges: The Science Question in Feminism and the Privilege
of Partial Perspective. Feminist Studies 14 (3): 575. doi: 10.2307/3178066.
Hess, Martin. 2004. ‘Spatial’ relationships? Towards a reconceptualization of embedded ness.
Progress in Human Geography 28 (2): 165–186. doi: 10.1191/0309132504ph479oa.
Jazeel, Tariq. 2016. Between area and discipline. Progress in Human Geography 40 (5): 649–667.
doi: 10.1177/0309132515609713.
Kellecioglu, Deniz. 2020. The deep coloniality of economics – a (r)evolutionary appraisal. Review
of African Political Economy 47 (165): 484–493. doi: 10.1080/03056244.2020.1763288.
Kitchin, Rob, Sandra Collins, and Dermot Frost. 2015. Funding models for Open Access digital data
repositories. Online Information Review 39 (5): 664–681. doi: 10.1108/OIR-01-2015-0031.
Kitchin, Rob, and Tracey P. Lauriault. 2019. Data and data infrastructures. In Digital Geographies,
ed. James Ash, Rob Kitchin and Agnieszka Leszczynski, 83–94. Los Angeles: Sage.
Kitchin, Rob, and Tracey P. Lauriault. 2015. Small data in the era of big data. GeoJournal 80 (4):
463–475. doi: 10.1007/s10708-014-9601-7.
Larivière, Vincent, Stefanie Haustein, and Philippe Mongeon. 2015. The Oligopoly of Academic
Publishers in the Digital Era. PLOS ONE 10 (6): e0127502. doi: 10.1371/journal.pone.0127502.
Larkin, Brian. 2013. The Politics and Poetics of Infrastructure. Annual Review of Anthropology 42
(1): 327–343. doi: 10.1146/annurev-anthro-092412-155522.
Marchais, Gauthier, Paulin Bazuzi, and Aimable Amani Lameke. 2020. ‘The data is gold, and we are
the gold-diggers’: whiteness, race and contemporary academic research in eastern DRC.
Critical African Studies 12 (3): 372–394. doi: 10.1080/21681392.2020.1724806.
Monbiot, George. 2018. Scientific publishing is a rip-off: We fund the research – it should be free.
The Guardian, September 13.
https://www.theguardian.com/commentisfree/2018/sep/13/scientific-publishing-rip-off-
taxpayers-fund-research (last accessed 25 August 2021).
Ndlovu-Gatsheni, Sabelo J. 2020). Decolonization, development and knowledge in Africa. London:
Routledge.
Ndlovu-Gatsheni, Sabelo J. 2017. Decolonising research methodology must include undoing its
dirty history. https://theconversation.com/decolonising-research-methodology-must-
include-undoing-its-dirty-history-83912 (last accessed 25 August 2021).
Nshobole, Judith B. 2020. ’Donor-researchers’ and ‘recipient-researchers’: Bridging the gap
between researchers from the Global North and Global South.
https://www.gicnetwork.be/donor-researchers-and-recipient-researchers-bridging-the-gap-
between-researchers-from-the-global-north-and-global-south/ (last accessed 25 August
2021).
Ouma, Stefan. 2015. Assembling export markets: The making and unmaking of global food
connections in West Africa. Chichester: Wiley-Blackwell.

20 Frontiers in African Digital Research
University of Bayreuth African Studies Online (10)
Peekhaus, Wilhelm. 2012. The Enclosure and Alienation of Academic Publishing: Lessons for the
Professoriate. tripleC: Communication, Capitalism & Critique. Open Access Journal for a Global
Sustainable Information Society 10 (2): 577–599. doi: 10.31269/triplec.v10i2.395.
Posada, Alejandro, and George Chen. 2018. Inequality in knowledge production: The integration of
academic infrastructure by big publishers. 22nd International Conference on Electronic
Publishing: OpenEdition Press.
Raworth, Kate. 2017. Doughnut economics: Seven ways to think like a 21st-century economist. White
River Junction, Vermont: Chelsea Green Publishing.
Sadowski, Jathan. 2020. The Internet of Landlords: Digital Platforms and New Mechanisms of
Rentier Capitalism. Antipode 52 (2): 562–580. doi: 10.1111/anti.12595.
Schurr, Carolin, Martin Müller, and Nadja Imhof. 2020. Who Makes Geographical Knowledge? The
Gender of Geography’s Gatekeepers. The Professional Geographer 72 (3): 317–331. doi:
10.1080/00330124.2020.1744169.
Schwartz-Shea, Peregrine, and Dvora Yanow. 2012. Interpretive research design: Concepts and
processes. New York: Routledge.
Shah, Alpa. 2019. Good research is vital. The Patriot, November 15.
https://thepatriot.in/2019/11/15/good-research-is-vital/ (last accessed 25 August 2021).
Smith, Linda Tuhiwai. 2012. Decolonizing Methodologies: Research and indigenous peoples, 2nd
edn. London: ZED Books.
Vega, Alejandra Manco; Shen, Cenyu; Armengou, Clara; Mitchell, Dominic et al. 2021. The directory
of open access journals. https://www.doaj.org/ (last accessed 25 August 2021).
Zienkowski, Jan. 2017. Reflexivity in the transdisciplinary field of critical discourse studies.
Palgrave Communications 3 (1). doi: 10.1057/palcomms.2017.7.